14: Steve Yegge/Gene Kim: 2 Hour Pair Programming Session! Captions Slightly Wrong? скачать в хорошем качестве

14: Steve Yegge/Gene Kim: 2 Hour Pair Programming Session! Captions Slightly Wrong?

Okay, let’s take Steve’s advice. Prompt: “this is so close! but in the captions, it's not clearing the previous caption correctly. bounding box problem?” We look at the generated code, and joke that it seems reasonable — this is obviously a joke, because the ffmpeg command is completely unreadable and difficult to comprehend. But let’s pull the slot machine handle again! Wait, before we do that, Steve has the best idea ever: Prompt: “please construct the ffmpeg command step by step and document, so we can follow along.” OMG, it is so much more readable! Each parameter is now on a separate line, and documented. So, so, so much better!!! And let’s hardcode the end time so that it only renders 20 seconds to make the runs go faster. Another pull of the slot machine handle at 3 min — while it runs, Steve asks what changed in the code. I answer, “It looks like it’s all new code.” Haha. Steve’s response: “Magic.” And then he posits why he is skeptical about coding agents at this time — he ponders whether you could just say, “Could you give an LLM a broad specification like, use ffmpeg, here’s an SRT file, here’s an input clip, generate the excerpt? Honestly, this was hard for a human—it wasn’t obvious and involved a lot of fumbling, restarts, and retries. If a person struggles with it, an agent is likely to as well. But I agree, over time, we’ll automate pieces of the process until you’re essentially the agent, just highly automated.” In the video, I agreed with him — and since then, I’ve thought more about it, and can describe with more logical rigor why. So much of the work in this session was concretizing and reifying the problem so that it could be solved. It was the incremental breaking down of problems so that the LLM could fix it. In a later conversation, Steve mentioned how over time, not only the leaves of the tree, but more of the layers above them (the parents of the leafs) are increasingly able to be delegated to an LLM. In other words, LLMs current excel at addressing well-defined, isolated tasks (the “leaves”), but over time, more of the bottom portions of the tree will be able to be solved by LLMs. Currently, LLMs lack the ability to abstract, iterate, and resolve ambiguities inherent in higher-level problem-solving. (For now, at least.) Without humans guiding the process, it would be like a blind, pinball machine, bouncing from idea to idea, never getting closer to the goal. Okay, we continue to philsophize until 5 minutes in — I’m wondering why it’s taking so long to render. (In hindsight, I think it’s because there’s multiple ffmpeg runs going — I later used “pkill” after each canceled run.) Steve mentions how he really wants to see what happens if you ask Claude to render the captions in a Quentin Tarantino font, hoping for yellow spaghetti-western font. Thirty seconds later, we’re looking at the new movie — the captions are on black background now (a big improvement as it’s more readable). And we're starting to put our fingers on what is wrong: the lines are rendering independently, for starters. (I will learn why later that evening.) One more pull of the slot machine — and we note that in two hours, we’re almost done building the tool. (Actually, looking at the timer, it’s only been one hour — the rest was chit-chat, joking, learning the tool, etc…) Prompt: “almost! it's rendering two lines of captions, which aren't updating correctly.| let's make it simpler, by showing only one line of caption. In Tarantino font. And use current style of composing ffmpeg command step by step, documented.” (And again, the generated code looks so different — comments are on the same line, instead of the line above it. Another great example of how non-deterministic they are!)